This tutorial is designed for beginners to learn Artificial Intelligence using Jetson Nano 2GB with JupyterLab. This is the easiest way to learn since the code is provided and we just need to run the program. With this, we don’t have to write and worry about the error of the codes. We can just learn by understanding the lines of codes one by one. So, let’s see what we can learn with it.
We need to install these software before we proceed with our learning. Below is the link for the software to donwload:
- Raspberry Pi Imager – To write the JetPack image into the SD card.
- PuTTy – To allows the use of SSH (Secure Shell) to access a remote computer.
Download the Jetpack SDK from this link. (Choose for Jetson Nano 2GB)
Below is the harware needed for this tutorial:
- NVIDIA Jetson Nano 2GB Dev Kit-No Wireless Adapter
- Jetson Nano 2GB Bundles
- CLiPtec 720P HD Wide-Angle USB Webcam with Mic
- USB Micro B Cable
Part 1: Getting Started with Jetson Nano 2GB
This video covers step by step how to start using the Jetson Nano 2GB. It’s absolutely for beginners. We are using serial communication to connect with Jetson Nano 2GB. We start with downloading and write the JetPack into the micro SD card. Then we will set up the Jetson Nano based on our preferences. After that, we will connect to the JupyterLab and test our camera.
Set The Data Directory
Create the directory:
mkdir -p ~/nvdli-data
Run The Container
For USB Camera:
sudo docker run --runtime nvidia -it --rm --network host \ --volume ~/nvdli-data:/nvdli-nano/data \ --device /dev/video0 \ nvcr.io/nvidia/dli/dli-nano-ai:v2.0.1-r32.5.0
For CSI Camera(Pi Camera):
sudo docker run --runtime nvidia -it --rm --network host \ --volume ~/nvdli-data:/nvdli-nano/data \ --volume /tmp/argus_socket:/tmp/argus_socket \ --device /dev/video0 \ nvcr.io/nvidia/dli/dli-nano-ai:v2.0.1-r32.5.0
Note: Please refer to DLI Getting Started with AI on Jetson Nano to know the suitable container tag for your current JetPack.
Connect to JupyterLab
Go to this link – http://192.168.55.1:8888 and login to the JupyterLab with password dlinano.